FLORA INDICA. 29 mately divided into lobes, and the amount of division seems (as was long ago pointed out by Scringe) to vary indefinitely. To such an extent, indeed, does the variation extend, that occasionally species very dissimilar in fruit are in a flowering state abso- lutely undistingnishable. This tendency to mutability of form is unfortunately not confined to the leaves, but extends to the size and degree of branching of the stem, to the size of the flowers, to the shape of the head of fruit and of the individual car- pels, and to the amount of pubescence; and in consequence the genus is in a state of complete chaos, the descriptions given in books being quite insufficient for the deter- mination of the species. Very frequently the diagnoses of the same plant given by different authors are quite irreconcilable, and the most different species are occasionally found iu herbaria under the same name. A careful examination of extensive suites of specimens from all parts of the world has convinced us that no single character,' except the colour of the flowers, is to be relied upon absolutely. The shape of the leaves is the least constant of all, and in four-fifths of the genus is undefinable in words; and even the shape of the style or beak of the achenia, which seems to be mainly relied on as a character, wiU, unless used with great caution, lead to very erroneous con- clusions, as straight and curved styles may be seen on the sanie specimen, frequently even iu the same capitulum. Nothing is more common than to find in botanical works that a newly-described species is " facile distinctus" by a certain character, which, if an extensive series of specimens be examined, wSl be found to be no character at all. At the same time we seek in vain in such works for any recog- nition of the great amount of variation to \vhich the different organs are subject, though the fact must be familiar to all careful observers of nature. And yet with this mass of ill-assorted descriptions in books, new species are almost daily being added to the "list, not a few being described without a knowledgejrf the ripe fruit. We believe that no greater boon could be conferred upon science than a careful series of observations on the amount of variation to which cultivated specimens of any com- mon Ranunculus are liable during a series of years. Sect. 1. BATEAOHIUM, DC.—Qarpella transverse rugosa. More* albi, petalorum ungue fiavo. 1. R. aqnatilis (L. fep. 781); fhitans, foliis submersis capillaceo- multifidis, emersis (dum adsu'nt) rotimdato-renifonnibus.—DC. Prod. i. 26; Don, in JRoyle, III 54; Schkcbt. Anim. Ran. 7; Ledeb. M. Moss. L '27; Torrey et Gray, Fl. N, Am. i. 15. B. divaricatus et fluitans, Ledeb. L c. E. peucedanifolius, AIL; SchkcU. Anim. I. c. B, Pantothrix et fluviatilis, Auct. HAB. Beluchistan! Afghanistan! Kashmir I Ladak usque ad 14,500 ped. alt.! Panjab Himalaya, Jacquemont! Romaon, alt. 5-12,000 ped.! in Tibetia Sikkimensi, alt. 17,000 ped.!; in India calida rarissizna: ad Saharunpur in planitie Gangetica superiore, &oyle'f—(WL per totam asstatem.) (t?. t>.) DISTRIB. Europa! usque ad Islandiara I Asia temperata usque ad CMnam! Tasmank! Abyssinia 1 Algeria! Teneriffa! America borealis temperata usque ad mare arcticum! Eerla aquatilis, in lacubus et aquis lente fluentibu« fluitaus, radicibus fibrosis. Caules sfopius eloiigati, graciles. Folia submersa petiokta, rariosve flewflia, 1-3 poDices Ipnga, circumscriptione rotundata, dissecta, segmeirtis capHlaceis; eixena (in specimmibus Indicis adhuc non observata) rotundato-reniformia, incisb-crenata, tri- loba vel tripartite Pedunculi oppositifolii, 1-flori. IHore* magnitadine Talderarii, diametro ^-1^-pollicares. Acltfnia in capitulum globosum collecta, ovali-oblonga, vix compressa.